From: Pablo Neira Ayuso <pablo@netfilter.org>
To: Jakub Kicinski <kuba@kernel.org>
Cc: netfilter-devel@vger.kernel.org, davem@davemloft.net,
netdev@vger.kernel.org, fw@strlen.de
Subject: Re: [PATCH net 3/9] netfilter: conntrack: collect all entries in one cycle
Date: Fri, 6 Aug 2021 16:59:54 +0200 [thread overview]
Message-ID: <20210806145954.GA1405@salvia> (raw)
In-Reply-To: <20210806062759.2acb5a47@kicinski-fedora-pc1c0hjn.dhcp.thefacebook.com>
[-- Attachment #1: Type: text/plain, Size: 457 bytes --]
On Fri, Aug 06, 2021 at 06:27:59AM -0700, Jakub Kicinski wrote:
> On Fri, 6 Aug 2021 13:52:01 +0200 Pablo Neira Ayuso wrote:
> > rcu_read_lock();
> >
> > nf_conntrack_get_ht(&ct_hash, &hashsz);
> > if (i >= hashsz)
> > - i = 0;
> > + break;
>
> Sparse says there is a missing rcu_read_unlock() here.
> Please follow up on this one.
Right.
I can squash this fix and send another PR or send a follow up patch.
Let me know your preference.
[-- Attachment #2: x.patch --]
[-- Type: text/x-diff, Size: 520 bytes --]
diff --git a/net/netfilter/nf_conntrack_core.c b/net/netfilter/nf_conntrack_core.c
index 3fdcf251ec1f..d31dbccbe7bd 100644
--- a/net/netfilter/nf_conntrack_core.c
+++ b/net/netfilter/nf_conntrack_core.c
@@ -1377,8 +1377,10 @@ static void gc_worker(struct work_struct *work)
rcu_read_lock();
nf_conntrack_get_ht(&ct_hash, &hashsz);
- if (i >= hashsz)
+ if (i >= hashsz) {
+ rcu_read_unlock();
break;
+ }
hlist_nulls_for_each_entry_rcu(h, n, &ct_hash[i], hnnode) {
struct nf_conntrack_net *cnet;
next prev parent reply other threads:[~2021-08-06 15:00 UTC|newest]
Thread overview: 14+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-08-06 11:51 [PATCH net 0/9] Netfilter fixes for net Pablo Neira Ayuso
2021-08-06 11:51 ` [PATCH net 1/9] netfilter: ipset: Limit the maximal range of consecutive elements to add/delete Pablo Neira Ayuso
2021-08-06 11:52 ` [PATCH net 2/9] netfilter: nf_conntrack_bridge: Fix memory leak when error Pablo Neira Ayuso
2021-08-06 11:52 ` [PATCH net 3/9] netfilter: conntrack: collect all entries in one cycle Pablo Neira Ayuso
2021-08-06 13:27 ` Jakub Kicinski
2021-08-06 14:59 ` Pablo Neira Ayuso [this message]
2021-08-06 15:06 ` Jakub Kicinski
2021-08-06 11:52 ` [PATCH net 4/9] netfilter: nfnetlink_hook: strip off module name from hookfn Pablo Neira Ayuso
2021-08-06 11:52 ` [PATCH net 5/9] netfilter: nfnetlink_hook: missing chain family Pablo Neira Ayuso
2021-08-06 11:52 ` [PATCH net 6/9] netfilter: nfnetlink_hook: use the sequence number of the request message Pablo Neira Ayuso
2021-08-06 11:52 ` [PATCH net 7/9] netfilter: nfnetlink_hook: Use same family as " Pablo Neira Ayuso
2021-08-06 11:52 ` [PATCH net 8/9] netfilter: conntrack: remove offload_pickup sysctl again Pablo Neira Ayuso
2021-08-06 11:52 ` [PATCH net 9/9] netfilter: nfnetlink_hook: translate inet ingress to netdev Pablo Neira Ayuso
2021-08-06 15:11 [PATCH net 0/9,v2] Netfilter fixes for net Pablo Neira Ayuso
2021-08-06 15:11 ` [PATCH net 3/9] netfilter: conntrack: collect all entries in one cycle Pablo Neira Ayuso
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20210806145954.GA1405@salvia \
--to=pablo@netfilter.org \
--cc=davem@davemloft.net \
--cc=fw@strlen.de \
--cc=kuba@kernel.org \
--cc=netdev@vger.kernel.org \
--cc=netfilter-devel@vger.kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).